LncRNA MAYA promotes iron overload and hepatocyte senescence through inhibition of YAP in non‐alcoholic fatty liver disease

Abstract Although recent evidence has shown that hepatocyte senescence plays a crucial role in the pathogenesis and development of non‐alcoholic fatty liver disease (NAFLD), the mechanism is still not clear. The purpose of this study was to investigate the signal transduction pathways involved in the senescence of hepatocyte, in order to provide a potential strategy for blocking the process of NAFLD. The results confirmed that hepatocyte senescence occurred in HFD‐fed Golden hamsters and PA‐treated LO2 cells as manifested by increased levels of senescence marker SA‐β‐gal, p16 and p21, heterochromatin marker H3K9me3, DNA damage marker γ‐H2AX and decreased activity of telomerase. Further studies demonstrated that iron overload could promote the senescence of hepatocyte, whereas the overexpression of Yes‐associated protein (YAP) could blunt iron overload and alleviate the senescence of hepatocyte. Of importance, depression of lncRNA MAYA (MAYA) reduced iron overload and cellular senescence via promotion of YAP in PA‐treated hepatocytes. These effects were further supported by in vivo experiments. In conclusion, these data suggested that inhibition of MAYA could up‐regulate YAP, which might repress hepatocyte senescence through modulating iron overload. In addition, these findings provided a promising option for heading off the development of NAFLD by abrogating hepatocyte senescence.


| INTRODUC TI ON
Non-alcoholic fatty liver disease (NAFLD) is the most frequent chronic liver disease in many areas of the world. According to the data published recently, the overall prevalence of NAFLD was 29.8% in China. 1 NAFLD, a multi-system disease related to metabolic disorders, is associated with liver injury 2 and may progress to advanced liver fibrosis, cirrhosis and even hepatocellular carcinoma if not controlled effectively. Up until now, there is no efficacious drugs for NAFLD, due to the fact that the molecular mechanism of NAFLD has not been elucidated completely. Therefore, it is essential for the discovery of effective target drugs to explore the novel therapeutic targets of NAFLD.
The senescence of hepatocyte plays an important role in the pathogenesis of NAFLD, and thus, hepatocyte senescence-related molecules have been becoming therapeutic targets for the treatment of NAFLD. 3 Cellular senescence, a consequence of replicative senescence, 4,5 is typically characterized by telomere dysfunction and shortening, up-regulation of ageing-related genes, accumulation of DNA damage and limits the proliferative potential of damaged cells through the induction of cell cycle arrest. 6,7 It was reported that hepatocyte senescence could promote hepatic steatosis in non-alcoholic fatty liver disease and elimination of ageing hepatocytes or inhibition of hepatocyte senescence might be effective strategies for the amelioration of NAFLD. 8 In recent years, a host of studies have shown that the pathogenesis of non-alcoholic fatty liver disease is frequently associated with iron overload. [9][10][11] As an organ for iron recycling, storage and iron-containing enzyme synthesis, liver plays a key role in the process of iron metabolism. The disorder of iron metabolism leads to excessive iron deposition in the liver of patients with NAFLD, which is usually called iron overload and thought to be involved in oxidative stress and mitochondrial dysfunction, two pathological factors associated with cellular senescence. [12][13][14][15] However, there is no evidence that iron overload can induce hepatocyte senescence in NAFLD.
At present, little is known about the specific signal pathways through which pathological iron overload promotes liver cell senescence. Yes-associated protein (YAP), a major downstream effector of the Hippo signalling pathway, has been reported to be involved in regulating cellular senescence. [16][17][18] However, the relationship between YAP and iron overload has not been explored. A recent study has reported that long non-coding RNA (lncRNA) MAYA (MST1/2-Antagonizing for YAP Activation) is involved in YAP activation. 19 LncRNAs, a novel category of non-coding RNAs, exert its regulatory functions through specific interactions with proteins, including transcriptional factors/coactivators, epigenetic modifiers and RNP complexes. 20 A few of top international academic journals showed that the regulation of lncRNA plays a considerable role in occurrence and progression of NAFLD. 21,22 In the early stage of this study, we verified that palmitic acid (PA) induced cellular senescence in LO2 cells. In addition, we found that the expression of MAYA increased in PA-treated LO2 cells. In the present study, we performed both in vivo and in vitro experiments to investigate the signal transduction pathways involved in hepatocyte senescence from the aspect of MAYA/YAP regulation of iron overload.  were adaptively fed for a week and then randomly divided into 2 groups (n = 8) as follows: Group 1 was the negative control (NC) in which hamsters were given a standard diet and hamsters in Group 2 were given a high-fat diet for 8 weeks. At the end of experiments, hamsters were killed after being anaesthetised by intraperitoneal injection with pentobarbital (50 mg/kg). A small portion of the liver was fixed in a 10% neutral buffered formalin for histological analysis, and the others were stored at −80℃ for further analyses, such as Oil Red O, SAβ-gal. Eight-week-old male C57BL/6J mice (18 ~ 20 g bodyweight, Nanjing Qinglongshan Experimental Animal Company) were randomly divided into 5 groups (n = 8) and given a standard diet in group1 and group 2 or a high-fat diet in the other groups after adaptive feeding for 1 week. After 10 weeks, mice in groups 2 and 3 were injected with sh-NC, and groups 4 and 5 were injected with sh-MAYA-1 and sh-MAYA-2 through the tail veil, respectively. The lentiviral3-GFP-shRNA specifically targeting MAYA was designed and synthesized by GenePharma (Shanghai, China), and nonspecific shRNA was used as negative control (sh-NC). Lentiviral vector (1 × 10 9 TU/ml) was injected into mice for 24 days (once per 8 days).

| | Experimental animal procedures
At the end of experiments, all the animals were killed after being anaesthetised by intraperitoneal injection with pentobarbital. Part of the liver was fixed in a 10% neutral buffered formalin for histological analysis and the others were stored at −80℃ for further analysis. The sequences of MAYA shRNA (mice) were as following: sh-MAYA#A: 5'-GCACATAGCTCTTGTCTTTAG-3'; sh-MAYA#B: 5'-GCCCAATTAAGGCAAGTAAGG-3'.

| | Cell culture
The human immortalized normal hepatocyte cell line LO2 cells (Cell Bank of Chinese Academy of Sciences, Shanghai, China) were cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Grand Island, NY, USA) with 10% foetal bovine serum and 1% penicillin and streptomycin, and grown in the 5% CO 2 incubator at 37℃.

| | Haematoxylin and eosin (H&E) staining and immunohistochemistry staining
Liver tissues fixed in 10% neutral buffered formalin were embedded in paraffin and cut into 4 μm thick slices. H&E staining and immunohistochemistry staining were performed as described in our previous study. 23

| | Oil Red O staining
Frozen section of liver tissues and cultured LO2 cells in 24-well plates were subjected to Oil Red O staining as previously described. 13

| | Iron content assay
The iron content of liver tissue was measured by Tissue Iron Content Colorimetric Assay Kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) following the manufacturer's instructions. Cellular

| | Immunofluorescence staining
Immunofluorescence staining of liver tissues or LO2 cells was performed in accordance with the standard procedure of previous report. 24 DAPI (Beyotime, Shanghai, China) stain solution was used to locate the nucleus of hepatocytes in vivo and in vitro.

| | SAβ-gal staining
SAβ-gal staining was performed based on the manufacturer's protocol. In brief, LO2 cells cultured in 24-well plates were washed with phosphate buffered saline (PBS) and fixed with 0.5ml fixative solution. The fixed cells were rinsed 2 times with PBS and incubated with 0.5ml/well of the β-galactosidase staining solution at 37℃ overnight in a dry incubator (no CO 2 ). Frozen sections of liver tissues were performed in accordance with that for LO2 cells.

| | Western blot analysis
LO2 cells or liver tissues were lysed in RIPA buffer containing protease inhibitors, and the protein concentration was quantified by using a BCA protein assay kit (Beyotime, Shanghai, China). The experimental procedure of Western blot was as described in our previous research. 23 Representative blots were from three independent experiments.

| | Statistical analysis
Results were presented as mean ±SEM, and statistical analysis was performed using GraphPad Prism 5.0 (GraphPad Software, San Diego, CA, USA). The significance of difference was determined by one-way analysis of variance with the post hoc Dunnett's test.
Values of P <.05 were considered statistically significant.

| RE SULTS
3.1 | | The senescence of hepatocyte occurred in the liver of hamsters with non-alcoholic fatty liver disease Golden hamster is an ideal model for NAFLD because the distribution and metabolism of blood lipids are similar to human. In order to clarify the hepatocyte senescence in NAFLD liver, we used LVG Golden Syrian hamsters to establish an in vivo model of NAFLD with a high-fat diet (HFD). 25 Photomicrographs of H&E staining showed that HFD feeding induced ballooning degeneration in liver tissues, accompanied by conspicuous cell swelling, cytoplasmic vacuolation and inflammatory infiltration ( Figure 1A). More accumulation of hepatic lipid droplets was observed in NAFLD hamsters than in control group as assessed by Oil Red O staining ( Figure 1B). SAβ-gal activity assay was used to detect cellular senescence and the results showed that SAβ-gal-positive hepatocytes increased significantly in the model group ( Figure 1C).
It is well known that telomere shortening and telomerase inhibition lead to replicative senescence. 26 Telomeric Repeat Binding Factor 1 and 2 (TRF1 and TRF2) are responsible for the major telomerase activity. 27,28 TRF1 negatively regulates telomere length, whereas TRF2 maintains telomere structure. 29,30 Therefore, we detected the mRNA levels of TERT, TRF1 and TRF2 by qRT-PCR and found that HFD feeding induced a significant down-regulation of TERT mRNA and up-regulation of TRF1 mRNA, whereas the level of TRF2 mRNA remained unchanged ( Figure 1D). Consistently, the analyses of senescence-associated genes showed that HFD treatment up-regulated the expression of senescence markers p16 and p21 ( Figure 1D). In addition, the protein levels of heterochromatin marker H3K9me3 and DNA damage marker γ-H2AX were increased in HFD feeding hamsters ( Figure 1E), providing further evidence that the cells enter into senescence. 31 Taken together, these results strongly illustrated that HFD-induced lipid accumulation may elicit hepatocyte senescence in vivo.

| | Iron overload mediated PA-induced cellular senescence in LO2 cells
To elucidate the correlation between NAFLD and iron overload, we first detected whether there is iron overload in in vivo and in vitro NAFLD models. The results showed that the iron contents in livers of NAFLD hamsters and PA-treated LO2 cells were significantly higher than those in the control groups ( Figure 3A and γ-H2AX when compared with PA-treated group ( Figure 3E).
Consistently, exposure to FAC further decreased the protein level of TERT and increased the protein level of TRF1 in PA-treated LO2 cells ( Figure 3F). These results indicated that iron overload mediated PA-induced cellular senescence in LO2 cells.

| | Overexpression of YAP inhibited iron overload and cellular senescence in PA-treated LO2 cells
Previous reports indicated that YAP is an important mediator of cellular senescence. 32, 33 We performed a series of experiments to detect whether YAP is related to iron overload and cell ageing in PA-treated LO2 cells. As illustrated in Figure 4A, the protein level of YAP was reduced under the PA treatment. On the other hand, we transfected LO2 cells with YAP CRISPR Activation Plasmid to up-regulate the expression of YAP ( Figure 4B). As expected, overexpression of YAP could alleviate iron overload in PA-treated LO2 cells ( Figure 4C) and ameliorate PA-induced intracellular lipid accumulation and cellular senescence ( Figure 4D). In accordance with these findings, analyses of Western blot and immunofluorescence staining displayed that overexpression of YAP led to a significant reversion of PA-induced up-regulation of senescence markers, including p16, p21, γ-H2AX and TRF1 ( Figure 4E,F). On the other hand, the effect of YAP expression plasmid on TERT from our trial was opposite to the above indicators  Figure 4F). Immunofluorescence staining provided consistent results by detecting the expression of senescence markers of γ-H2AX and H3K9me3 ( Figure 4G). These data suggested that YAP mediated iron overload and cellular senescence in PA-treated LO2 cells.

| | MAYA drove iron overload and hepatocyte senescence via down-regulation of YAP protein
LncRNAs are involved in regulating RNA, DNA and protein interactions and further modulates intracellular signalling pathways. 34 LncRNA MAYA (MAYA), also known as MNX1-AS1 (LOC645249), (binds MST1, LLGL2, SAV1, MOB1, NSUN6 and RASSF1) has been reported to mediate the Hippo/YAP pathway by methylating Hippo/MST1 at Lys59. 19 Our previous data showed that the mRNA level of MAYA was enhanced in PA-treated LO2 cells ( Figure 5A).
We here assumed that inhibition of MAYA might stimulate the expression of YAP, which in turn reduced hepatocyte senescence and ultimately improved NAFLD. To confirm this assumption, we down-regulated the level of MAYA by shRNA silencing, and the re-  Figure 6A, 6B), with the levels of senescence marker p16, p21 and γ-H2AX decreased evidently ( Figure 6C). In addition, inhibition of MAYA reduced the protein level of TRF1, whereas the protein level of TERT was elevated significantly ( Figure 6D). Taken together, these data collectively indicated that knock-down of MAYA

| D ISCUSS I ON
NAFLD, characterized by excessive accumulation of hepatic lipids, is closely linked to non-alcoholic steatohepatitis (NASH), cirrhosis and hepatocellular carcinoma. 35 Although the pathogenesis of NAFLD have been extensively studied, the mechanisms associated with cellular senescence are not characterized. 36 Therefore, understanding the molecular mechanisms of cellular senescence will help find ways to improve NAFLD. At the beginning of this study, we verified The cell cycle arrest of senescent cells is mainly manifested by increased expression of cell cycle inhibitors, such as p16 and p21. 4 γ-H2AX is regarded as a marker of early DNA damage, which is a characteristic of cellular senescence. 39,40 Besides, the formation of H3K9me3-containing heterochromatin foci is associated with the progression of cellular senescence. 41 Prior studies have shown that mutations in members of the telomerase complex and accumulating of DNA damage exist in NAFLD. 41,42 We herein confirmed that hepatocyte senescence in in vivo and in vitro NAFLD models by detecting the above series of ageing indicators. Moreover, these results would help us further clarify its potential mechanism.
Iron, an effective pro-oxidant, participates in the transport of oxygen and the process of tissue respiration and maintains the body's homeostasis. Iron overload could induce an imbalance of the homoeostatic mechanisms, which contributes to liver damage resulting from glucose and lipid metabolism disorders. 43 In addition, iron overload damages cells and causes cellular senescence through oxidative stress. 13 However, it is not clear whether iron overload can regulate hepatocyte senescence in NAFLD. In this report, we confirmed that iron deposition existed in the liver of HFD-treated hamsters and PA-treated LO2 cells. Then, we used iron overload agonist FAC to clarify whether the regulation of iron overload can reduce the senescence of hepatocyte in NAFLD. We found that FAC cells. Moreover, senescence marker (SAβ-Gal activity, p16, p21), DNA damage marker (γ-H2AX), the telomere length and telomerase activity (TRF1, TERT) were significantly different from only PA-or FAC-treated LO2 cells, suggesting that PA-induced cellular senescence was aggravated by iron overload. Therefore, the inhibition of iron overload could be an effective target for the reduction in hepatocyte senescence in NAFLD. Certainly, additional study will be needed to illustrate the detailed mechanism of iron overload regulating hepatocyte senescence. In conclusion, the aggregate data in this study indicated that suppression of MAYA had the capacity to reduce cellular senescence and ameliorate NAFLD, which might be related to the regulation of YAP and subsequent amelioration of iron overload ( Figure 7H). However, these results do not rule out the possible involvement of any other signalling mechanisms in the hepatocyte senescence of NAFLD. Our findings suggested a novel lncRNAmediated approach to improve NAFLD through modulation of hepatocyte senescence.

ACK N OWLED G EM ENTS
This study was supported by the National Natural Science

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest. Conceptualization (equal).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.